CN113095543A - Distribution method and system for airport parking space and taxiways - Google Patents
Distribution method and system for airport parking space and taxiways Download PDFInfo
- Publication number
- CN113095543A CN113095543A CN202110227757.4A CN202110227757A CN113095543A CN 113095543 A CN113095543 A CN 113095543A CN 202110227757 A CN202110227757 A CN 202110227757A CN 113095543 A CN113095543 A CN 113095543A
- Authority
- CN
- China
- Prior art keywords
- taxiway
- stand
- allocation
- sample
- parking space
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 94
- 238000009826 distribution Methods 0.000 title claims description 32
- 230000009471 action Effects 0.000 claims abstract description 47
- 230000008569 process Effects 0.000 claims abstract description 24
- 238000012549 training Methods 0.000 claims description 45
- 238000004088 simulation Methods 0.000 claims description 24
- 238000013528 artificial neural network Methods 0.000 claims description 16
- 238000005457 optimization Methods 0.000 claims description 15
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 9
- 238000003860 storage Methods 0.000 claims description 8
- 238000010276 construction Methods 0.000 claims description 5
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 4
- 230000010006 flight Effects 0.000 description 13
- 239000011159 matrix material Substances 0.000 description 8
- 238000011160 research Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 5
- 238000013461 design Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 238000011156 evaluation Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 230000007704 transition Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 230000002068 genetic effect Effects 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000005728 strengthening Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 230000033001 locomotion Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
- 238000010845 search algorithm Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/40—Business processes related to the transportation industry
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2111/00—Details relating to CAD techniques
- G06F2111/08—Probabilistic or stochastic CAD
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Biomedical Technology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Tourism & Hospitality (AREA)
- Molecular Biology (AREA)
- General Business, Economics & Management (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Marketing (AREA)
- Development Economics (AREA)
- Probability & Statistics with Applications (AREA)
- Entrepreneurship & Innovation (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Medical Informatics (AREA)
- Quality & Reliability (AREA)
- Algebra (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
Abstract
The invention provides a method and a system for allocating airport parking spaces and taxiways, wherein the method comprises the following steps: constructing total resource state information of a stand and a taxiway in a target airport based on a Markov decision process model, and converting the total resource state information into a resource state two-dimensional view; and inputting the resource state two-dimensional view into a trained parking space and taxiway allocation strategy network to obtain a parking space and taxiway allocation action strategy, and allocating the parking space and the taxiway in the target airport according to the parking space and taxiway allocation action strategy. The invention realizes the efficient cooperative dynamic allocation of two resources of the parking space and the taxiway under the large-scale hub airport by constructing the layered Markov decision process model suitable for the dynamic joint allocation of the parking space and the taxiway, improves the operating efficiency of the airport, and saves the energy and the operating cost.
Description
Technical Field
The invention relates to the technical field of airport operation optimization, in particular to an allocation method and system for airport parking spaces and taxiways.
Background
Along with the rapid development of economy, air transportation becomes one of people's main trip modes, and the civil aviation industry develops rapidly to having offered huge challenge to the civil aviation operation management, the airport is as the air traffic pressure that the beginning-to-end point of flow bore constantly increases, and scene operating efficiency is influenced, needs urgent research to promote the new theory and the technique of scene operating efficiency.
In the operation process of the airport, the allocation result of a Gate directly influences the allocation scheme of personnel and materials, and plays an important role in ensuring the safety and the efficient operation of the airport. The Taxiway (taxi) directly connected with the stand is a passage for the departure flight to enter and exit the stand, and the optimal allocation of the Taxiway can effectively save energy and reduce the operation cost. The joint allocation of the parking places and the taxiways has a crucial influence on the scene operation management of the airport and the trip experience of passengers.
Although the joint assignment problem of the parking spaces and the taxiways is very hot, the existing research on the problem is still in a preliminary stage, most researchers only take the relevant indexes of the taxiways as one of the factors for evaluating the assignment of the parking spaces, for example, a multi-target airport parking space assignment problem model taking sliding collision avoidance as a safety constraint, an assignment scheme such as maximum optimized target selection, maximum parking space occupancy rate, minimum passenger walking distance and the like are established, but the problem is essentially scheduled for single resources, the assignment of the parking spaces and the taxiways is not carried out simultaneously, and heuristic algorithms such as tabu search, ant colony algorithm or genetic algorithm and the like are mostly adopted when the problem is solved. And a small part of research really realizes the allocation of two resources under the condition of small data volume, establishes the optimization target of the minimum walking time of passengers and the minimum taxi-in and taxi-out time of aircrafts, and solves the problems by combining a plurality of heuristic algorithms such as a genetic algorithm, a tabu search algorithm and the like, thereby realizing the parking space allocation and the taxi path planning. However, in general, the research on the existing problems is still in the preliminary stage. Therefore, there is a need for a method and system for assigning airport parking spaces and taxiways to solve the above problems.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a method and a system for allocating airport parking spaces and taxiways.
The invention provides an allocation method for airport parking spaces and taxiways, which comprises the following steps:
constructing total resource state information of a stand and a taxiway in a target airport based on a Markov decision process model, and converting the total resource state information into a resource state two-dimensional view;
inputting the resource state two-dimensional view into a trained parking space and taxiway allocation strategy network to obtain a parking space and taxiway allocation action strategy, and allocating the parking space and the taxiway in the target airport according to the parking space and taxiway allocation action strategy, wherein the trained parking space and taxiway allocation strategy network is obtained by training a parking space strategy network and a taxiway strategy network according to a sample resource state two-dimensional view, and the parking space strategy network and the taxiway strategy network are neural networks.
According to the method for allocating the airport parking spaces and the taxiways, which is provided by the invention, the trained parking space and taxiway allocation strategy network is obtained by training through the following steps:
constructing sample overall resource state information according to sample physical resource state information, sample logic resource state information and sample resource occupation time state information corresponding to the sample flight schedule, and converting the sample overall resource state information into a sample resource state two-dimensional view;
and respectively training the parking place strategy network and the taxiway strategy network through the sample resource state two-dimensional view to obtain the trained parking place and taxiway allocation strategy network.
According to the method for allocating the airport parking space and the taxiways, which is provided by the invention, the parking space strategy network and the taxiway strategy network are trained respectively through the two-dimensional view of the sample resource state to obtain the trained parking space and taxiway allocation strategy network, and the method comprises the following steps:
respectively inputting the two-dimensional view of the resource state of the sample into a stand strategy network and a taxiway strategy network, and performing plot simulation based on a Monte Carlo method to obtain a stand state sample and a taxiway state sample corresponding to each simulation moment, wherein the stand state sample comprises a general resource state sample of a stand and a taxiway, a stand action selection sample and a stand immediate reward sample, and the taxiway state sample comprises a general resource state sample of a stand and a taxiway, a taxiway action selection sample and a taxiway immediate reward sample;
and respectively training and updating parameters of the stand strategy network and the taxiway strategy network according to the stand state sample and the taxiway state sample, and if preset training conditions are met, obtaining a trained stand and taxiway distribution strategy network.
According to the invention, the method for allocating the airport parking space and the taxiways further comprises the following steps:
taking the maximized near-position allocation rate and the minimized super-far-position allocation rate as the allocation optimization target of the parking positions, and constructing an immediate parking position reward sample based on the immediate reward of the taxiway conflict condition;
and taking the minimized taxiway conflict rate as an allocation optimization target of the taxiway, and constructing and obtaining an immediate taxiway reward sample.
According to the method for allocating the airport stand and the taxiway, the parameters of the stand strategy network and the parameters of the taxiway strategy network are trained and updated respectively according to the stand state sample and the taxiway state sample, and before the trained stand and taxiway allocation strategy network is obtained if the preset training conditions are met, the method further comprises the following steps:
constructing constraint conditions according to flight attribute information and stop allocation rule information, wherein the constraint conditions comprise flight and stop matching constraint conditions and taxiway and stop matching constraint conditions;
updating the stand distribution probability in each simulation moment based on the flight and stand matching constraint conditions and the stand strategy network parameters, and acquiring a stand action selection sample according to the updated stand distribution probability;
and updating the taxiway allocation probability at each simulation moment based on the matching constraint condition of the taxiway and the stand and the taxiway strategy network parameters, and acquiring a taxiway action selection sample according to the updated taxiway allocation probability.
According to the method for allocating the airport stand and the taxiway, the parameters of the stand strategy network and the taxiway strategy network are trained and updated according to the stand state samples and the taxiway state samples, and the method comprises the following steps:
and training and updating parameters of the parking place and taxiway distribution strategy network according to the parking place state sample and the taxiway state sample by a strategy gradient algorithm.
According to the method for allocating the airport parking spaces and taxiways, after the constraint condition is constructed according to the flight attribute information and the parking space allocation rule information, the method further comprises the following steps:
and acquiring the corresponding stand or taxiway for allocation by a roulette method based on the updated stand allocation probability or the updated taxiway allocation probability.
The present invention also provides a distribution system for airport stands and taxiways, comprising:
the resource state two-dimensional view construction module is used for constructing the total resource state information of the parking spaces and the taxiways in the target airport based on a Markov decision process model and converting the total resource state information into a resource state two-dimensional view;
and the parking space and taxiway allocation module is used for inputting the resource state two-dimensional view into a trained parking space and taxiway allocation strategy network to obtain a parking space and taxiway allocation action strategy, and allocating the parking space and the taxiway in the target airport according to the parking space and taxiway allocation action strategy, wherein the trained parking space and taxiway allocation strategy network is obtained by training a parking space strategy network and a taxiway strategy network according to a sample resource state two-dimensional view, and the parking space strategy network and the taxiway strategy network are neural networks.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor when executing the program implementing the steps of any of the methods for airport stands and taxiways as described above.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, performs the steps of any of the methods for airport parking space and taxiway allocation as described above.
According to the method and the system for allocating the parking spaces and the taxiways of the airport, the layered Markov decision process model suitable for dynamic combined allocation of the parking spaces and the taxiways is constructed, so that efficient collaborative dynamic allocation of two resources, namely the parking spaces and the taxiways, under a large-scale hub airport is realized, the operating efficiency of the airport is improved, and the energy and the operating cost are saved.
Drawings
In order to more clearly illustrate the technical solutions of the present invention or the prior art, the drawings needed for the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method for assigning airport stands and taxiways in accordance with the present invention;
FIG. 2 is a two-dimensional view of the resource status of the stand and taxiway provided by the present invention;
FIG. 3 is a schematic diagram of an architecture of a parking place and taxiway allocation policy network according to the present invention;
FIG. 4 is a schematic structural view of a distribution system for airport stands and taxiways provided by the present invention;
fig. 5 is a schematic structural diagram of an electronic device provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Most of the existing airport resource allocation management systems are lack of automatic allocation engines with high feasibility, and although certain algorithm assistance is provided in the actual operation of airports, the final decision still depends on manual work, so that the pressure of workers is huge in the early and late peak and large-area flight delay. Therefore, the reasonable combined allocation scheme of the parking spaces and the taxiways for multiple resources can greatly improve the efficiency of the airport, form the operation integration of the airport and greatly save manpower, material resources and financial resources.
In addition, the existing research joint distribution scheme is combined, and the following defects exist in terms of design innovation of a problem model and real realization of joint dynamic distribution of the airplane parking spaces and taxiways under a large hub airport: firstly, traditional problem model curing, wherein the existing research almost totally carries out multi-objective optimization problem model design and adopts a common heuristic algorithm to solve, so that the distribution result lacks flexibility and adaptability to disturbance flights; secondly, the effect of experimental data is limited, the problem of joint allocation of parking spaces and taxiways is an NP-hard problem, the complexity of calculation time of the NP-hard problem increases exponentially with the increase of the sizes of airports and flights, and for a large hub airport, it is extremely difficult to obtain an optimal solution in a short time, and other intelligent algorithms are usually needed. However, the flight amount that can be handled by the existing implementation method is only within hundreds of flights, and the allocated flight plan can be selected only a few or more than ten hours in one day, and cannot adapt to the huge data amount of a large hub airport, so that the researched joint allocation method is difficult to be really applied to the operation of an actual airport. Therefore, the design of the combined distribution method for the machine parking place and the taxiway under the large hub airport has high practical significance.
The Markov Decision Process (MDP) framework has strong abstraction and flexibility, can be applied to many different problems in many ways, introduces the concept of time-stepping, does not require fixed real-time intervals, and can be used to refer to any stage of Decision and action. In the problem of dynamic combined allocation of airport stand and taxiway resources, flight arrival is discrete, limited and sequential, and the design of a hierarchical MDP framework is innovative and feasible. Furthermore, the invention adopts a deep reinforcement learning method, and can effectively solve the problem of joint allocation of the parking space and the taxiway.
Fig. 1 is a schematic flow chart of an allocation method for airport stands and taxiways according to the present invention, and as shown in fig. 1, the present invention provides an allocation method for airport stands and taxiways, which includes:
The invention establishes a layered Markov decision process model on the basis of the traditional optimization problem model to complete the conversion from the traditional model to the layered MDP model. Since flights arrive discretely according to the time sequence, the arrival of the flights needs to be monitored by a stall and taxiway allocation strategy algorithm at each moment, an environment state view, namely a resource state two-dimensional view, is constructed according to the current resource state and the flight queue state, corresponding stall and taxiway joint allocation actions are executed according to the environment state, and allocation results are counted in a period of time to achieve the predicted optimization target.
Further, the MDP model of the present invention may be represented by a quintuple as
Wherein the content of the first and second substances,
the state space is represented by a representation of,
representing an action space, and P representing a state transition probability matrix; r ═ R (s, a) } represents an immediate reward vector generated according to the action and state, and
j denotes the evaluation function of the policy, which is denoted as J (θ) when the policy is represented by a function or neural network with a parameter of pi. In the invention, the state space of the layered MDP structure is the states of the aircraft stand and the taxiway of the airport, and the strategy evaluation functions are two, namely J (theta) and J '(theta'), which are respectively used for representing the respective neural network strategies of the aircraft stand and the taxiway. Aiming at different neural networks, the action spaces are different, namely the mapping of selecting the stand a and the mapping of selecting the taxiway a ', and the total action space a ═ (a, a') is obtained by combination; depending on the selected action and state, different neural networks will produce different rewards r and r ', which in combination result in an overall immediate reward r ″, r + r', thereby taking advantage of the differencesRespective strategies for respectively enforcing the stand and taxiway are awarded. The state transition probability matrix is the probability matrix after the airport stand and the taxiway state change. In the invention, the intelligent agent senses the initial environment, implements actions according to the current strategy, enables the initial environment to enter a new state under the influence of the actions, and feeds back a reward to the intelligent agent; the agent then takes a new policy, based on the new state, to continue interacting with the environment.
Because the dimension of the state space of the combined dynamic collaborative allocation system of the parking spaces and the taxiways in the large-scale hub airport is extremely high, the influence factors of state change are complex, the state transition probability matrix P is difficult to analyze, and the MDP problem is extremely difficult to solve by adopting a theoretical method, the invention designs a layered strategy network training architecture on the basis of the establishment of a layered MDP model, solves the MDP problem of the combined allocation of the parking spaces and the taxiways by adopting a layered DRL architecture, constructs a two-layer strategy neural network, and verifies the effectiveness, the high-efficiency processing capability and the applicability to the complex large-scale hub airport by a simulation experiment.
In the invention, the environmental state S of the parking space and the taxiway of the airport is sensed by the intelligent agent at any momenttAfter the environment state is converted into a resource state two-dimensional view, a machine position strategy network with a parameter theta represented by a deep neural network and a taxiway strategy network with a parameter theta' are respectively input to decide a machine position distribution result atAnd a taxiway assignment result at', to obtain an immediate reward corresponding to the two allocation resultsrtAnd rt'; further, the environmental status of the airport stand and taxiway is updated to St+1. It should be noted that, when the policy network is allocated to the stand and the taxiway, the scenario simulation is performed by using the monte carlo method, so that a large number of policy network training sample tracks are obtained:
and obtaining a stand state sample and a taxiway state sample at each simulation moment, respectively using the two state samples for training different strategy networks, and training and updating strategy network parameters theta and theta' by using a strategy gradient method. After the strategy network training is finished, flight is allocated in real time by using the strategy network, the operation speed of cooperative allocation of the parking space and the taxiway can be greatly increased, the problem solving efficiency is improved, the problem solving method can adapt to huge data volume of a large hub airport, and the capability of dynamic cooperative allocation of the parking space and the taxiway under a large number of flights and a long-time flight plan is provided.
The method for allocating the parking spaces and the taxiways of the airport provided by the invention realizes the efficient and cooperative dynamic allocation of two resources of the parking spaces and the taxiways under a large-scale hub airport by constructing a layered Markov decision process model suitable for the dynamic joint allocation of the parking spaces and the taxiways, improves the operating efficiency of the airport, and saves energy and operating cost.
On the basis of the above embodiment, the trained stand and taxiway allocation strategy network is trained by the following steps:
constructing sample overall resource state information according to sample physical resource state information, sample logic resource state information and sample resource occupation time state information corresponding to the sample flight schedule, and converting the sample overall resource state information into a sample resource state two-dimensional view;
and respectively training the parking place strategy network and the taxiway strategy network through the sample resource state two-dimensional view to obtain the trained parking place and taxiway allocation strategy network.
On the basis of the above embodiment, the method further includes:
the maximized near-line position allocation rate and the minimized super-far line position allocation rate are used as the allocation optimization target of the machine halt positions, and meanwhile, in order to better meet the cooperativity of the allocation of two resources (namely the maximized near-line position allocation rate and the minimized super-far line position allocation rate), when the immediate rewarding sample of the machine halt positions is constructed, the immediate rewarding of the taxiway conflict condition is considered;
and taking the minimized taxiway conflict rate as an allocation optimization target of the taxiway, and constructing and obtaining an immediate taxiway reward sample.
In the invention, the combined dynamic collaborative allocation problem of the parking place and the taxiway with large data volume is modeled into a hierarchical MDP model, and the MDP model is constructed before the training of the parking place and taxiway allocation strategy network. First, an immediate reward for the MDP model is designed, the present invention to maximize the near slot allocation rate and minimize the far slot allocation rate F (Y)1,Y2) And minimizing taxiway conflict rate F2(Y2) For the optimization goal, where Y1Decision variables, Y, representing stand2Decision variables representing taxiways:
wherein, yijWhen the number is 1, the flight i is allocated to the stand j, j belongs to the M, otherwise yij＝0；yizWhen 1, it means that flight i is assigned to taxiway Z, Z ∈ Z. The decomposed optimization target corresponds toSum of immediate rewards for a track for different policy networks:
wherein γ ═ 1, represents the influence coefficient of the future immediate reward; r ist＝Rj+RzImmediate reward for parking place policy network, RjFor reward or punishment value of selected stand, RzThe reward and punishment value of the selected taxiway; r ist′＝RzAn immediate reward representing a network of taxiway policies achieves the optimization goal of cooperative distribution through the decomposition and combination of the immediate rewards. It should be noted that the airport flight data information in the present invention includes M stops, N flights, and Z taxiways, and three types of stops are considered: when the flight is allocated to the near seat, the passenger can directly check on the airplane through the gallery bridge; when the flight is distributed to a far airport, the passenger needs to go to and from the terminal building and the airport through the ferry; when a flight is assigned to an extra far gate, it indicates that no suitable near gate and far gate can be assigned, defined as a gate assignment failure. In addition, two taxiway conflict scenarios are also considered: the method comprises the following steps of (1) carrying out head-to-head conflict and rear-end conflict, wherein the head-to-head conflict refers to that two flights are opposite in direction on the same taxiway in the same time and do not keep enough safety distance; and the rear-end collision indicates that two flights have the same direction on the same taxiway at the same time but do not meet the safe distance condition.
Further, overall resource state construction of the stand and taxiways is performed. Specifically, the input states of the stand strategy network and the taxiway strategy network are the same and are the total resource state information of the stand and the taxiway, and the resource state comprises three kinds of state information: physical resource state information, logical resource state information, and resource occupancy time state information. The physical resource state information is used for representing the actual occupation situation of the stand and the taxiway resource from the current time T to the time T + T, and is mathematically represented as an 0/1 matrix:
wherein, a·j,t1 means that gate j is occupied at time t, a'·z,t1 indicates that the taxiway z is occupied at time t.
The logical resource status information indicates the available parking spaces and taxiways for each of the future L flights and indicates those parking spaces and taxiways that may be parked and driven for a particular arriving flight in compliance with the actual operating regulations for the airport. In the invention, the logic operation rule considers the international and domestic attributes of the flight, the affiliated airline company, the flight task category of the flight, the model of the flight and the fixed matching rule of the parking space and the taxiway, and simultaneously, the occupied parking space and the taxiway at the current moment are removed, and the logic operation rule is expressed in a mathematical form as follows:
wherein, blj,t1 represents that the stand j meets the parking condition of the scheduled flight l at the time t; dlz,t1, the taxiway z is indicated as a taxiway that the flight l can pass through at time t.
The resource occupation time status information is used to describe the respective parking time of the future L flights, i.e. the difference between the departure time and the arrival time, and is expressed as follows:
Ct＝(t1 t2 … tL)T；
and finally, splicing the three types of information to obtain the total resource state information of the stand and the taxiway at the current time t:
St＝St′＝(At Bt Ct)；
further, in order to enable the parking space and the taxiway allocation policy network to better identify and train, the invention converts the matrix form corresponding to the total resource state information into a two-dimensional view, and then the two-dimensional view information is respectively input into the layered policy network, fig. 2 is a resource state two-dimensional view of the parking space and the taxiway provided by the invention, as shown in fig. 2, the first part is a Physical resource state view (Physical resource state), the second and fourth parking spaces can be seen to be occupied, and the first and second taxiways are used simultaneously; the second part is a Logical resources status view (Logical resources state), and it can be seen that the airport is actually occupied at the current moment by the second and fourth stands, and unfilled stands represent stands and taxiway resources that do not meet the actual operating rules of the airport, wherein each cell in the physical resources status view and the Logical resources status view represents the length of unit time occupied. It should be noted that the extra-far aircraft position and the taxiways of the extra-far aircraft position are always available, such as the rightmost area of the aircraft stand and the rightmost area of the taxiways in the logic resource state view in fig. 2, and the squares of the areas are all filled; the last column (Residnce time) represents the resource occupancy time status information for the future flight.
Finally, the motion space is modeled. Specifically, when a flight arrives at an airport according to a flight schedule, a stand and a corresponding taxiway are assigned, so that the action space is M × Z, and the selection actions of the stand policy network and the taxiway policy network are respectively:
at＝aj；
at′＝az′；
wherein, ajIndicating the selected stand j, az' denotes the selected taxiway z.
On the basis of the above embodiment, the training of the stand policy network and the taxiway policy network respectively through the sample resource state two-dimensional view to obtain the trained stand and taxiway allocation policy network includes:
respectively inputting the two-dimensional view of the resource state of the sample into a stand strategy network and a taxiway strategy network, and performing plot simulation based on a Monte Carlo method to obtain a stand state sample and a taxiway state sample corresponding to each simulation moment, wherein the stand state sample comprises a general resource state sample of a stand and a taxiway, a stand action selection sample and a stand immediate reward sample, and the taxiway state sample comprises a general resource state sample of a stand and a taxiway, a taxiway action selection sample and a taxiway immediate reward sample;
and respectively training and updating parameters of the stand strategy network and the taxiway strategy network according to the stand state sample and the taxiway state sample, and if preset training conditions are met, obtaining a trained stand and taxiway distribution strategy network.
On the basis of the above embodiment, before the parameters of the stand policy network and the taxiway policy network are trained and updated according to the stand state sample and the taxiway state sample, respectively, and if a preset training condition is met, a trained stand and taxiway allocation policy network is obtained, the method further includes:
constructing constraint conditions according to flight attribute information and stop allocation rule information, wherein the constraint conditions comprise flight and stop matching constraint conditions and taxiway and stop matching constraint conditions;
updating the stand distribution probability in each simulation moment based on the flight and stand matching constraint conditions and the stand strategy network parameters, and acquiring a stand action selection sample according to the updated stand distribution probability;
and updating the taxiway allocation probability at each simulation moment based on the matching constraint condition of the taxiway and the stand and the taxiway strategy network parameters, and acquiring a taxiway action selection sample according to the updated taxiway allocation probability.
Fig. 3 is a schematic structural diagram of a parking space and taxiway allocation policy network provided by the present invention, which can be referred to fig. 3, in the present invention, the parking space and taxiway scene state s (overall resource state information) of an airport are abstracted as a two-dimensional view of resource state; and then the resource state view is respectively input into a stand strategy network and a taxiway strategy network, and the flight and stand matching constraint condition B is modeled into action selection of the stand strategy network and the taxiway strategy network. Further, after the resource state two-dimensional view is input into the strategy network, a stall position action selection a ' is obtained, a corresponding stall position reward and punishment value r ' is obtained, then the taxi channel action selection a ' is obtained according to the taxi channel strategy network and the constraint condition D of matching the taxi channel and the stall position, a corresponding taxi channel reward and punishment value r ' is obtained, then the r ' is used for strengthening the selection of the taxi channel strategy network, the r ' and the r ' are combined to obtain the collaborative distribution reward and punishment value r, the r is used for strengthening the selection of the stall position strategy network, and finally the stall position and the taxi channel resource state of the target airport are updated according to the comprehensive action a ═ (a ', a '). In the present invention, the taxiway and stand matching constraint D is expressed as:
wherein d isij1 means that taxiway i can be matched with stand j, otherwise dij＝0。
Further, the training process of the parking space and taxiway allocation strategy network of the invention is specifically explained, and the steps are as follows:
step 201, initializing strategy network training parameters, including flight schedule sample number E, training iteration times I, number K of parallel simulation plots of each flight schedule sample in each training, and maximum time step number T of each plot simulation;
step 202, initializing attributes of the stand and information of stand distribution rules;
step 203, reading and initializing flight schedule samples. Specifically, reading flight schedule samples, and if the number of the read flight schedule samples is less than E, superposing random time disturbance on each flight in the schedule to generate a new flight schedule sample until the number of the flight schedule samples is equal to E;
step 204, acquiring flight and stop matching constraint information and taxiway and stop matching constraint information according to the flight attribute information, the stop attribute and distribution rule information;
step 205, setting structural parameters of a hierarchical strategy neural network, and initializing the strategy network by using coefficients such as random weight, bias and the like;
step 206, initializing a strategy network training loop variable i to 1, e to 1, and k to 1;
step 207, starting the strategy network training of the ith round;
step 208, selecting an e-th flight schedule sample;
step 209, performing K episode simulations according to the current flight schedule sample to obtain two tracks of each episode simulation:
step 210, calculating the state value in each plot simulation:
step 211, updating strategy network coefficients theta and theta' by adopting a Monte Carlo REINFORCE with baseline method;
step 212, judging whether all the E flight schedule samples in the round of training are simulated, if so, entering step 213, otherwise, returning to step 208, if not, E is E + 1;
in step 213, it is determined whether the I-th round of training is completed, if so, the process proceeds to step 214, otherwise, I is I +1, and e is 1, and the process returns to step 207.
Step 214, storing the trained strategy network piθAnd pi'θ′And finishing the training.
On the basis of the above embodiment, the training and updating the parameters of the stand strategy network and the taxiway strategy network according to the stand state samples and the taxiway state samples includes:
and training and updating parameters of the parking place and taxiway distribution strategy network according to the parking place state sample and the taxiway state sample by a strategy gradient algorithm.
On the basis of the above embodiment, after the constraint condition is constructed according to the flight attribute information and the gate allocation rule information, the method further includes:
and acquiring the corresponding stand or taxiway for allocation by a roulette method based on the updated stand allocation probability or the updated taxiway allocation probability.
In the present invention, the scenario simulation process is specifically explained, and the steps are as follows:
step 301, when t is equal to 0, initializing parking space and taxiway occupation states, and constructing a parking space and taxiway state matrix
(conversion to two-dimensional view);
step 302, starting simulation of the t time step;
step 303, judging whether a flight exists in the flight state, if so, entering step 304, otherwise, constructing a parking space and taxiway state matrix, setting 0, and jumping to step 311;
step 304, will
Input stopAnd (p) calculating the probability p (p) that the flight to be distributed is distributed to each stand1,p2,…,pM)；
Step 305, adding constraint conditions (flight and stand matching constraint conditions) for the distribution probability of stands, setting the probability of illegal stands to 0, and obtaining the updated probability p of standsg；
Step 306, according to pgSelecting a practical assigned parking space for the flight to be assigned by roulette
Step 307, will
Inputting a taxiway policy network, and calculating to obtain the probability p' ═ p of each taxiway allocated to flights to be allocated1′,p2′,…,pZ′)；
308, increasing the matching constraint conditions of the taxiways and the parking positions for the probability distribution of the taxiways, and setting 0 to the probability of the illegal taxiway to obtain the updated probability p of the taxiwayt′；
Step 309, according to pt' the method of roulette is adopted to select a actually-allocated taxiway for the flight to be allocated
Step 310, calculate immediate reward rt kAnd rt k′；
Step 311, t equals t + 1;
step 312, judging whether T is greater than T, if yes, ending the current plot simulation, otherwise, performing step 313;
313, updating the two-dimensional view of the states of the stand and the taxiways according to the distribution results of the stand and the taxiways
Then will have already beenThe flight that assigns the gate and taxiway exits the resource status view and reads in subsequent flights, returning to step 303.
In the invention, a trained parking space and taxiway allocation strategy network is tested through a test sample set, the average value of each performance evaluation result of all samples is counted, the method is compared with the parking space and taxiway cooperative allocation problem method solved by the existing optimization software Gurobi, and the effectiveness and the efficiency are compared, so that the effectiveness and the high efficiency of the method are proved; compared with a method for solving the problem of cooperative allocation of the stand and the taxiway by a heuristic algorithm Greedy, the method disclosed by the invention researches the influence of different cooperative coefficients on different resource decision-making schedules, and proves that the method can adjust the weight among multiple optimization targets and has better adaptability. It should be noted that, the testing process steps of the present invention may refer to the training process and the simulation process, and are not described herein again.
Fig. 4 is a schematic structural diagram of an allocation system for airport stands and taxiways provided by the present invention, and as shown in fig. 4, the present invention provides an allocation system for airport stands and taxiways, which includes a resource state two-dimensional view construction module 401 and a stand and taxiway allocation module 402, wherein the resource state two-dimensional view construction module 401 is configured to construct overall resource state information of stands and taxiways in a target airport based on a markov decision process model, and convert the overall resource state information into a resource state two-dimensional view; the parking space and taxiway allocation module 402 is configured to input the resource state two-dimensional view into a trained parking space and taxiway allocation policy network to obtain a parking space and taxiway allocation action policy, and allocate a parking space and a taxiway in the target airport according to the parking space and taxiway allocation action policy, where the trained parking space and taxiway allocation policy network is obtained by training a parking space policy network and a taxiway policy network from a sample resource state two-dimensional view, and the parking space policy network and the taxiway policy network are neural networks.
The distribution system for the airport parking spaces and the taxiways provided by the invention realizes the efficient and cooperative dynamic distribution of two resources, namely the parking spaces and the taxiways under a large-scale hub airport, by constructing the layered Markov decision process model suitable for the dynamic combined distribution of the parking spaces and the taxiways, thereby improving the operating efficiency of the airport and saving energy and operating cost.
The system provided by the present invention is used for executing the above method embodiments, and for the specific processes and details, reference is made to the above embodiments, which are not described herein again.
Fig. 5 is a schematic structural diagram of an electronic device provided in the present invention, and as shown in fig. 5, the electronic device may include: a processor (processor)501, a communication interface (communication interface)502, a memory (memory)503 and a communication bus 504, wherein the processor 501, the communication interface 502 and the memory 503 are communicated with each other through the communication bus 504. The processor 501 may invoke logic instructions in the memory 503 to perform a method for airport stand and taxiway assignment, the method comprising: constructing total resource state information of a stand and a taxiway in a target airport based on a Markov decision process model, and converting the total resource state information into a resource state two-dimensional view; inputting the resource state two-dimensional view into a trained parking space and taxiway allocation strategy network to obtain a parking space and taxiway allocation action strategy, and allocating the parking space and the taxiway in the target airport according to the parking space and taxiway allocation action strategy, wherein the trained parking space and taxiway allocation strategy network is obtained by training a parking space strategy network and a taxiway strategy network according to a sample resource state two-dimensional view, and the parking space strategy network and the taxiway strategy network are neural networks.
In addition, the logic instructions in the memory 503 may be implemented in the form of software functional units and stored in a computer readable storage medium when the logic instructions are sold or used as independent products. Based on such understanding, the technical solution of the present invention may be embodied in the form of a software product, which is stored in a storage medium and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-only memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the method for airport parking and taxiway allocation provided by the above methods, the method comprising: constructing total resource state information of a stand and a taxiway in a target airport based on a Markov decision process model, and converting the total resource state information into a resource state two-dimensional view; inputting the resource state two-dimensional view into a trained parking space and taxiway allocation strategy network to obtain a parking space and taxiway allocation action strategy, and allocating the parking space and the taxiway in the target airport according to the parking space and taxiway allocation action strategy, wherein the trained parking space and taxiway allocation strategy network is obtained by training a parking space strategy network and a taxiway strategy network according to a sample resource state two-dimensional view, and the parking space strategy network and the taxiway strategy network are neural networks.
In yet another aspect, the present invention also provides a non-transitory computer-readable storage medium having stored thereon a computer program that, when executed by a processor, is implemented to perform the method for airport stand and taxiway allocation provided by the above embodiments, the method comprising: constructing total resource state information of a stand and a taxiway in a target airport based on a Markov decision process model, and converting the total resource state information into a resource state two-dimensional view; inputting the resource state two-dimensional view into a trained parking space and taxiway allocation strategy network to obtain a parking space and taxiway allocation action strategy, and allocating the parking space and the taxiway in the target airport according to the parking space and taxiway allocation action strategy, wherein the trained parking space and taxiway allocation strategy network is obtained by training a parking space strategy network and a taxiway strategy network according to a sample resource state two-dimensional view, and the parking space strategy network and the taxiway strategy network are neural networks.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding in mind, the above-described technical solutions may be embodied in the form of a software product, which can be stored in a computer-readable storage medium such as ROM/RAM, magnetic disk, optical disk, etc., and includes instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods described in the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A method for assigning airport stands and taxiways, comprising:
constructing total resource state information of a stand and a taxiway in a target airport based on a Markov decision process model, and converting the total resource state information into a resource state two-dimensional view;
inputting the resource state two-dimensional view into a trained parking space and taxiway allocation strategy network to obtain a parking space and taxiway allocation action strategy, and allocating the parking space and the taxiway in the target airport according to the parking space and taxiway allocation action strategy, wherein the trained parking space and taxiway allocation strategy network is obtained by training a parking space strategy network and a taxiway strategy network according to a sample resource state two-dimensional view, and the parking space strategy network and the taxiway strategy network are neural networks.
2. The method for airport stand and taxiway assignment according to claim 1, wherein the trained stand and taxiway assignment strategy network is trained by:
constructing sample overall resource state information according to sample physical resource state information, sample logic resource state information and sample resource occupation time state information corresponding to the sample flight schedule, and converting the sample overall resource state information into a sample resource state two-dimensional view;
and respectively training the parking place strategy network and the taxiway strategy network through the sample resource state two-dimensional view to obtain the trained parking place and taxiway allocation strategy network.
3. The method of claim 2, wherein the training of the stand policy network and the taxiway policy network to obtain the trained stand and taxiway allocation policy network by the two-dimensional view of the sample resource states comprises:
respectively inputting the two-dimensional view of the resource state of the sample into a stand strategy network and a taxiway strategy network, and performing plot simulation based on a Monte Carlo method to obtain a stand state sample and a taxiway state sample corresponding to each simulation moment, wherein the stand state sample comprises a general resource state sample of a stand and a taxiway, a stand action selection sample and a stand immediate reward sample, and the taxiway state sample comprises a general resource state sample of a stand and a taxiway, a taxiway action selection sample and a taxiway immediate reward sample;
and respectively training and updating parameters of the stand strategy network and the taxiway strategy network according to the stand state sample and the taxiway state sample, and if preset training conditions are met, obtaining a trained stand and taxiway distribution strategy network.
4. The method for assigning airport stands and taxiways according to claim 3, further comprising:
taking the maximized near-position allocation rate and the minimized super-far-position allocation rate as the allocation optimization target of the parking positions, and constructing an immediate parking position reward sample based on the immediate reward of the taxiway conflict condition;
and taking the minimized taxiway conflict rate as an allocation optimization target of the taxiway, and constructing and obtaining an immediate taxiway reward sample.
5. The method of claim 3, wherein before the parameters of the stand policy network and the taxiway policy network are updated according to the stand state samples and the taxiway state samples, respectively, and a trained stand and taxiway allocation policy network is obtained if a preset training condition is met, the method further comprises:
constructing constraint conditions according to flight attribute information and stop allocation rule information, wherein the constraint conditions comprise flight and stop matching constraint conditions and taxiway and stop matching constraint conditions;
updating the stand distribution probability in each simulation moment based on the flight and stand matching constraint conditions and the stand strategy network parameters, and acquiring a stand action selection sample according to the updated stand distribution probability;
and updating the taxiway allocation probability at each simulation moment based on the matching constraint condition of the taxiway and the stand and the taxiway strategy network parameters, and acquiring a taxiway action selection sample according to the updated taxiway allocation probability.
6. The method of airport stand and taxiway assignment according to claim 3, wherein said training updates to parameters of the stand policy network and the taxiway policy network based on the stand state samples and the taxiway state samples comprise:
and training and updating parameters of the parking place and taxiway distribution strategy network according to the parking place state sample and the taxiway state sample by a strategy gradient algorithm.
7. The method for airport stand and taxiway assignment as claimed in claim 5, wherein after said constructing constraints based on flight attribute information and stand assignment rule information, the method further comprises:
and acquiring the corresponding stand or taxiway for allocation by a roulette method based on the updated stand allocation probability or the updated taxiway allocation probability.
8. A distribution system for airport stands and taxiways, comprising:
the resource state two-dimensional view construction module is used for constructing the total resource state information of the parking spaces and the taxiways in the target airport based on a Markov decision process model and converting the total resource state information into a resource state two-dimensional view;
and the parking space and taxiway allocation module is used for inputting the resource state two-dimensional view into a trained parking space and taxiway allocation strategy network to obtain a parking space and taxiway allocation action strategy, and allocating the parking space and the taxiway in the target airport according to the parking space and taxiway allocation action strategy, wherein the trained parking space and taxiway allocation strategy network is obtained by training a parking space strategy network and a taxiway strategy network according to a sample resource state two-dimensional view, and the parking space strategy network and the taxiway strategy network are neural networks.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor, when executing the computer program, carries out the steps of the method for allocation of airport stands and taxiways according to any one of claims 1 to 7.
10. A non-transitory computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for allocation of airports stands and taxiways according to one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110227757.4A CN113095543B (en) | 2021-03-01 | 2021-03-01 | Distribution method and system for airport stand and taxiway |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110227757.4A CN113095543B (en) | 2021-03-01 | 2021-03-01 | Distribution method and system for airport stand and taxiway |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113095543A true CN113095543A (en) | 2021-07-09 |
| CN113095543B CN113095543B (en) | 2024-01-12 |
Family
ID=76667868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110227757.4A Active CN113095543B (en) | 2021-03-01 | 2021-03-01 | Distribution method and system for airport stand and taxiway |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113095543B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116933662A (en) * | 2023-09-14 | 2023-10-24 | 中国民用航空飞行学院 | Airport stand allocation method and device, storage medium and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106875054A (en) * | 2017-02-15 | 2017-06-20 | 民航成都信息技术有限公司 | A kind of flight resource dynamic early-warning method based on expert knowledge library |
| CN106981221A (en) * | 2017-03-24 | 2017-07-25 | 北京航空航天大学 | The airport break indices method and system decomposed based on time space dimension |
| CN107230392A (en) * | 2017-06-08 | 2017-10-03 | 大连交通大学 | Optimizing distribution method based on the hub aircraft gate for improving ACO algorithms |
| CN109147396A (en) * | 2018-08-23 | 2019-01-04 | 北京工业大学 | The distribution method and device of airport aircraft gate |
| CN111552178A (en) * | 2020-04-23 | 2020-08-18 | 桂林电子科技大学 | Method for controlling waiting release of aircraft stand with controllable repeat request time interval |
-
2021
- 2021-03-01 CN CN202110227757.4A patent/CN113095543B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106875054A (en) * | 2017-02-15 | 2017-06-20 | 民航成都信息技术有限公司 | A kind of flight resource dynamic early-warning method based on expert knowledge library |
| CN106981221A (en) * | 2017-03-24 | 2017-07-25 | 北京航空航天大学 | The airport break indices method and system decomposed based on time space dimension |
| CN107230392A (en) * | 2017-06-08 | 2017-10-03 | 大连交通大学 | Optimizing distribution method based on the hub aircraft gate for improving ACO algorithms |
| CN109147396A (en) * | 2018-08-23 | 2019-01-04 | 北京工业大学 | The distribution method and device of airport aircraft gate |
| CN111552178A (en) * | 2020-04-23 | 2020-08-18 | 桂林电子科技大学 | Method for controlling waiting release of aircraft stand with controllable repeat request time interval |
Non-Patent Citations (2)
| Title |
|---|
| NIE TONGTONG等: "A greedy algorithm based on joint assignment of airport gates and taxiways in large hub airports", HIGH TECHNOLOGY LETTEＲS, vol. 26, no. 4, pages 417 - 423 * |
| 刘君强等: "基于航班延误的机场滑行道停机位分配模型研究", 武汉理工大学学报, vol. 44, no. 4, pages 653 - 657 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116933662A (en) * | 2023-09-14 | 2023-10-24 | 中国民用航空飞行学院 | Airport stand allocation method and device, storage medium and electronic equipment |
| CN116933662B (en) * | 2023-09-14 | 2023-12-15 | 中国民用航空飞行学院 | Airport stand allocation method and device, storage medium and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113095543B (en) | 2024-01-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Mirheli et al. | A consensus-based distributed trajectory control in a signal-free intersection | |
| Yu et al. | MIP-based heuristics for solving robust gate assignment problems | |
| Bennell et al. | Airport runway scheduling | |
| CN104751681B (en) | Statistical learning model based gate position allocation method | |
| Samà et al. | Air traffic optimization models for aircraft delay and travel time minimization in terminal control areas | |
| Liu et al. | Robust assignment of airport gates with operational safety constraints | |
| Soltani et al. | An eco-friendly aircraft taxiing approach with collision and conflict avoidance | |
| CN114358446B (en) | Robust optimization method for airport resource scheduling | |
| Yin et al. | Joint apron-runway assignment for airport surface operations | |
| de Arruda Junior et al. | A new airport collaborative decision making algorithm based on deferred acceptance in a two-sided market | |
| CN114282823A (en) | Vehicle scheduling method and device, storage medium and electronic equipment | |
| Luo et al. | Research on situation awareness of airport operation based on Petri nets | |
| Bagamanova et al. | Reducing airport environmental footprint using a disruption-aware stand assignment approach | |
| CN113095543A (en) | Distribution method and system for airport parking space and taxiways | |
| Yang et al. | Stochastic scheduling of ground movement problem integrated with taxiway routing and gate/stand allocation | |
| Nogueira et al. | Using ant algorithm to arrange taxiway sequencing in airport | |
| Fernandes et al. | Optimization of the waiting time and makespan in aircraft departures: A real time non-iterative sequencing model | |
| Chen et al. | Multi-agent planning and coordination for automated aircraft ground handling | |
| Gołda et al. | Elements of the model positioning of aircraft on the apron | |
| Zhao et al. | Research on airport multi-objective optimization of stand allocation based on simulated annealing algorithm | |
| Das et al. | Deep Learning Based Negotiation Strategy Selection for Cooperative Conflict Resolution in Urban Air Mobility | |
| Zhang et al. | Research on flight first service model and algorithms for the gate assignment problem | |
| Khoury et al. | Evaluation of general-purpose construction simulation and visualization tools for modeling and animating airside airport operations | |
| Liang et al. | Sequence optimization of departure flights considering ferry scheduling | |
| Bubalo et al. | Reducing airport emissions with coordinated pushback processes: A case study |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |